Towards Software Defined Measurement with Open vSwitches: Designs, Implementation, and Evaluation
Zili Zha, An Wang, Yang Guo, Songqing Chen
Network measurement is critical for various network operations and management but has been often constrained by the available resources in the traditional network devices. Recent advances in Software-Defined Networking (SDN) have enabled flexible and programmable network measurement, which is referred to as Software Defined Measurement (SDM). A promising trend for SDM is to conduct network traffic measurement on widely deployed Open vSwitches (OVS) in data centers. However, little attention has been paid to the design options for conducting traffic measurement on the OVS. In this study, we set to explore different design choices and investigate the corresponding trade-offs among resource consumption, measurement accuracy, implementation complexity, and impact on switching speed. For this purpose, we explore the design space and empirically design and implement four different measurement schemes in OVS, by either closely integrating forwarding and measurement functions into a pipeline, or decoupling them into parallel operations. We further implement another emerging approach that is independent from OVS and include in our comparative study. Through extensive experiments and comparisons, we quantitatively show the various trade-offs (e.g., among the metrics of throughput, latency, CPU overhead, memory overhead) that the different schemes strike to balance, and demonstrate the feasibility of instrumenting OVS with monitoring capabilities. These results provide valuable insights into which design will best serve different measurement and monitoring needs.
, Wang, A.
, Guo, Y.
and Chen, S.
Towards Software Defined Measurement with Open vSwitches: Designs, Implementation, and Evaluation, IEEE Transactions on Cloud Computing, [online], https://doi.org/10.1109/TCC.2022.3181890, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=932942
(Accessed October 2, 2022)